Ectopic and Visceral Fat Deposition in Lean and Obese Patients With Type 2 Diabetes

Background Type 2 diabetes (T2D) and obesity are associated with nonalcoholic fatty liver disease, cardiomyopathy, and cardiovascular mortality. Both show stronger links between ectopic and visceral fat deposition, and an increased cardiometabolic risk compared with subcutaneous fat. Objectives This study investigated whether lean patients (Ln) with T2D exhibit increased ectopic and visceral fat deposition and whether these are linked to cardiac and hepatic changes. Methods Twenty-seven obese patients (Ob) with T2D, 15 Ln-T2D, and 12 normal-weight control subjects were studied. Subjects underwent cardiac computed tomography, cardiac magnetic resonance imaging (MRI), proton and phosphorus MR spectroscopy, and multiparametric liver MR, including hepatic proton MRS, T1- and T2*-mapping yielding “iron-corrected T1” [cT1]. Results Diabetes, with or without obesity, was associated with increased myocardial triglyceride content (p = 0.01), increased hepatic triglyceride content (p = 0.04), and impaired myocardial energetics (p = 0.04). Although cardiac structural changes, steatosis, and energetics were similar between the T2D groups, epicardial fat (p = 0.04), hepatic triglyceride (p = 0.01), and insulin resistance (p = 0.03) were higher in Ob-T2D. Epicardial fat, hepatic triglyceride, and insulin resistance correlated negatively with systolic strain and diastolic strain rates, which were only significantly impaired in Ob-T2D (p < 0.001 and p = 0.006, respectively). Fibroinflammatory liver disease (elevated cT1) was only evident in Ob-T2D patients. cT1 correlated with hepatic and epicardial fat (p < 0.001 and p = 0.01, respectively). Conclusions Irrespective of body mass index, diabetes is related to significant abnormalities in cardiac structure, energetics, and cardiac and hepatic steatosis. Obese patients with T2D show a greater propensity for ectopic and visceral fat deposition.

T ype 2 diabetes (T2D) and obesity are both associated with nonalcoholic fatty liver disease (NAFLD), cardiomyopathy (1,2), and increased cardiovascular mortality (3,4). The incidence of T2D continues to increase, driven predominantly by the obesity epidemic. Although obesity is likely to be a strong contributor to diabetic cardiomyopathy (5), many patients with diabetic cardiomyopathy have normal body mass index (BMI), suggesting that diabetes and obesity may have different mechanisms by which they mediate cardiovascular change and that diabetic cardiomyopathy may occur in patients with T2D without obesity. Furthermore, evidence suggests that distribution of excess fat is an important determinant of cardiovascular risk, and ectopic and visceral adiposity confer a much higher risk than subcutaneous adiposity (6,7).
Ectopic and visceral fat storage may be linked to insulin resistance, and it is widely known that insulin resistance is the strongest predictor of development of diabetes (8).
Increasing evidence points to a strong association between insulin resistance and nonischemic heart failure (9), although there are differing opinions regarding whether this relationship is of a protective or pathological nature (10)(11)(12).
Thus, the presence of ectopic and visceral fat deposition in patients with T2D even in the absence of a global increase in total body fat may potentially play a significant role in this association. Assessing body composition is, therefore, likely to be more important in patients with T2D than simple metrics of obesity.
Liver fat is considered a key feature of ectopic fat associated with dysfunctional adipose tissue and visceral fat deposition (13), and there is also growing interest in the imaging of epicardial adipose tissue as a proxy measure of visceral fat.
Epicardial adipose tissue, a form of visceral fat, may affect the underlying myocardium by secreting a wide range of adipokines (14). Furthermore, excess liver fat has been shown to be accompanied by cardiac structural and functional changes (15).  (16), allows noninvasive quantification of liver fat and identification of the presence of hepatic fibroinflammatory disease with a high diagnostic accuracy (16).
Myocardial energetic compromise is an important feature of both the diabetic (17) and the nondiabetic obese heart (5     LGE (Exclusion of myocardial scarring)

Visit 3
Suitability of patients with type 2 diabetes (T2D) was assessed during the first hospital visit. Those patients who consented to have a cardiac computed tomography (CT) scan were then invited for the second hospital visit. The third hospital visit included magnetic resonance imaging (MRI) and magnetic resonance spectroscopy scans (3T).
Multiparametric liver MRI included proton magnetic resonance spectroscopy ( 1 H-MRS) for hepatic triglyceride; T 1 and T 2 * mapping yielded iron-corrected T 1 (cT 1 ). This was followed by cardiac magnetic resonance, which included cine imaging to assess left ventricular (LV) volumes, mass, and ejection fraction; myocardial tagging for assessment of peak circumferential systolic strain and diastolic strain rate; cardiac 1 H-MRS for myocardial triglyceride; and late gadolinium enhancement (LGE) imaging for exclusion of myocardial scarring. Control subjects underwent identical MRI protocols. 31 P-MRS ¼ phosphorus magnetic resonance spectroscopy; ECG ¼ electrocardiogram; HLA ¼ horizontal long axis; PCr/ATP ¼ myocardial phosphocreatine to adenosine triphosphate concentration ratio; SA ¼ short axis. LIVER MRI. The liver multiparametric MR protocol has been previously described (16). MR scans were performed using a 3-T scanner (Tim Trio, Siemens).    Table 3. Similar to cardiac steatosis, diabetes, even in the absence of obesity, was associated with hepatic steatosis (hepatic triglyceride content in Ln-T2D vs. control subjects; p ¼ 0.044); however, hepatic

DISCUSSION
This study demonstrates for the first time that diabetes, even in the absence of obesity, is associated with significant cardiac structural and metabolic abnormalities, whereas significant functional changes, such as reductions in peak systolic strain and diastolic strain rates, are only evident in obese patients with diabetes. Furthermore, we show that those patients with diabetes who are also obese have higher   (23,24); moreover, we also demonstrate that there is an association between fibroinflammatory liver disease and insulin resistance in patients with diabetes.
The correlation of systolic strain and diastolic strain rates with hepatic and epicardial fat and insulin resistance suggests a link between these in patients with diabetes. However, the causality of these relationships will need to be investigated in future studies.
Finally, as is widely known, the spectrum of NAFLD ranges from fatty liver alone to nonalcoholic steatohepatitis (25). We show here that diabetes, even in the absence of obesity, is associated with hepatic steatosis at the mild end of the liver disease spectrum, but not with significant fibroinflammatory liver disease. Importantly, using multiparametric liver imaging, we show that significant NAFLD and nonal- glucose metabolism (26). Multiple studies support the concept that insulin resistance is prompted, and sustained by, dysregulated fat tissue (27)(28)(29). In addition, insulin resistance and ectopic adiposity are associated with an even greater cardiovascular risk (30,31), and obese subjects with T2D are at high risk of developing ectopic adiposity (32). There are many molecular mechanisms that may contribute to the association between insulin resistance and nonischemic cardiomyopathy (9). These include metabolic inefficiency (33) There is evidence of a role for the sympathetic nervous system in the relationship between insulin and hypertension in obese patients with hypertension (51). We did not demonstrate any significant difference in resting heart rates or the systolic blood pressure between the 2 diabetes groups to suggest an enhanced adrenergic drive in the obese group, but we did not assess circulating catecholamine levels.
Finally, the observational nature of our findings ox.ac.uk.